基于样本扩张灰色关联分析的光伏出力预测

发布时间：2018-05-17 02:04

本文选题：灰色关联度分析 + 样本扩张　；参考：《太阳能学报》2017年11期

【摘要】：光伏出力准确预测是光伏并网安全运行的重要基础,样本容量增大、计及多影响因素能有效提高光伏出力预测精度。以小时段为单位建立一种基于样本扩张灰色关联分析的光伏发电短期出力预测模型,扩张有限的样本容量,能分析多因素影响。首先分析影响光伏出力的多种因素,通过灰色关联度分析的方法对样本进行分析,得到扩张最优相似小时段样本;通过遗传算法对BP神经网络的权值和阈值进行优化,并对神经网络进行训练;最后进行光伏出力预测。该文所建立的预测模型有效扩张了样本容量,提高了突变天气时预测准确度,有一定应用价值。
[Abstract]:Accurate prediction of photovoltaic force is an important basis for the safe operation of photovoltaic grid. The sample size is increased and the precision of photovoltaic force prediction can be effectively improved by taking into account many factors. Based on the grey correlation analysis of sample expansion, a prediction model for short-term output of photovoltaic power generation is established in the unit of hours. The model can be used to analyze the influence of multiple factors by expanding the limited sample size. Firstly, the factors affecting photovoltaic output are analyzed, and the sample is analyzed by grey correlation analysis, and the sample in the extended optimal similar hour segment is obtained. The weight and threshold of BP neural network are optimized by genetic algorithm, and the weight and threshold of BP neural network are optimized by genetic algorithm. The neural network is trained and the photovoltaic force prediction is carried out at last. The prediction model established in this paper can effectively expand the sample size and improve the prediction accuracy of abrupt weather. It has certain application value.
【作者单位】：东南大学电气工程学院;
【分类号】：TM615

【相似文献】